3526
2. Stylianou E, Saklatvala J. Interleukin-1. Int J Biochem Cell Biol 1998;30:1075–9. 3. Tang C, Sula MJ, Bohnet S, Rehman A, Taishi P, Krueger JM. Interleukin-1 induces CREB-binding protein (CBP) mRNA in brain and the sequencing of rat CBP. Brain Res Mol Brain Res 2005;137:213–22. 4. Sambol EB, Ambrosini G, Geha RC, Kennealey PT, DeCarolis P, O’Connor R, et al. Flavopiridol targets c-KIT transcription and induces apoptosis in gastrointestinal stromal tumor cells. Cancer Res 2006;66:5858–66.
DOI 10.1002/art.38817
Reply To the Editor: We are appreciative that Chung took the time to carefully read our article and bring up interesting and salient points for discussion. We agree that some genes do not seem to “fit” with the conclusion that flavopiridol inhibits the induction of inflammatory mediator genes. However, we believe that we accurately represented the data by stating that most genes were not induced by IL-1 in the presence of flavopiridol, and that we were transparent by also showing genes that were exceptions and by including the raw data as supplemental information. In our study, the response of some genes was not expected, especially the responses of CCR5 and STAT1, both of which are implicated in inflammatory arthritis and osteoarthritis (1–3). For example, CDK-9 inhibition in the presence of interferon-␥ inhibited STAT1 activation and downstream effects in other cell types (4). It would be reasonable to expect a similar response to IL-1. Although messenger RNA levels (e.g., of STAT1 and CCR5) may not always correspond to protein levels or biologic activity, protein levels or biologic activities were not examined, because that was beyond the scope of the current study. We agree that further analysis of these genes would better explain the exact relationship of IL-1 and CDK-9/flavopiridol with these particular inflammatory mediators, especially in the context of cartilage and chondrocytes. There are several additional points to consider when evaluating these unexpected findings. First, in our experiments we used a 5-hour time point for the PCR array analysis. This time point was chosen because it yielded high-magnitude responses to the IL-1 treatment, but it does not necessarily preclude the possibility of secondary responses that are not regulated in the same CDK-9–dependent manner. Second, primary human cell strains and human cartilage explants remain largely untested with respect to CDK-9/flavopiridol and the regulation of primary response gene transcriptional elongation. These cells and tissues may not respond exactly as the other cell lines or tissues in the studies referenced by Chung. Finally, there is considerable variation in the response of human cartilage and isolated chondrocytes to IL-1. In our study, we repeated all experiments on samples from least 3 different human donors, indicating that the results shown are somewhat reproducible for both the expected responses and the unexpected responses of genes such as CCR5 and STAT1.
LETTERS
In summary, most of the inflammatory genes in the PCR array panel responded similarly, although there were a few outliers that had unexpected responses. Given the important roles of some of these outliers in arthritis, their relationship with the CDK-9–flavopiridol axis is fertile ground for further investigations. Dominik R. Haudenschild, PhD Jasper H. N. Yik, PhD University of California, Davis Sacramento, CA 1. Kasperkovitz PV, Verbeet NL, Smeets TJ, van Rietschoten JG, Kraan MC, Kraan T, et al. Activation of the STAT1 pathway in rheumatoid arthritis. Ann Rheum Dis 2004;63:233–9. 2. Lee YH, Bae SC, Song GG. Association between the chemokine receptor 5 ⌬32 polymorphism and rheumatoid arthritis: a metaanalysis. Mod Rheumatol 2013;23:304–10. 3. Qin SX, Rottman JB, Myers P, Kassam N, Weinblatt M, Loetscher M, et al. The chemokine receptors CXCR3 and CCR5 mark subsets of T cells associated with certain inflammatory reactions. J Clin Invest 1998;101:746–54. 4. Terashima T, Hague A, Kajita Y, Takeuchi A, Nakagawa T, Yokochi T. Flavopiridol inhibits interferon-␥-induced nitric oxide production in mouse vascular endothelial cells. Immunol Lett 2012;148:91–6.
DOI 10.1002/art.38812
Fibromyalgia syndrome and small-fiber neuropathy: comment on the article by Caro and Winter To the Editor: I read with interest the recent article by Caro and Winter (1), in which they reported decreased epidermal nerve fiber density (ENFD) in a series of 41 patients with fibromyalgia (FM), in accordance with other recent studies suggesting the involvement of small nerve fibers in FM as shown by reduced ENFD (2–4) and abnormal function of silent C-fiber nociceptors (5). These findings strengthen the concept that pain in FM syndrome is actually of the neuropathic type and point to the role of peripheral components, apparently opposite to the current opinion of FM as a condition characterized by central pain amplification (6). Unfortunately, Caro and Winter reported only that the overall mean ENFD was lower in patients compared with controls, without detailing the findings in individual cases. This is a crucial point, because the involvement of small nerve fibers seems to be limited to a subset of patients with FM. In previous studies by Oaklander et al (3) and Giannoccaro et al (4), only a minority of patients had reduced intraepidermal ENFD in ¨ ¸ceyler et al (2) reported abnormalities skin biopsy specimens. U in the function and morphology of small fibers in all 25 FM patients they studied. However, direct evidence of damage to small fibers was provided by reduced intraepidermal ENFD in only 16 patients, whereas the remaining patients had abnormal results on other tests (quantitative sensory testing and pain-related evoked potentials) that are not strictly specific for small fibers but rather investigate the whole length of the somatosensory system. In contrast, Serra et al (5) demon-
LETTERS
3527
strated hyperexcitability of silent (type 1B) C-fiber nociceptors, resembling that in small-fiber neuropathy (SFN), in the majority of patients with FM (23 of 30). The role of SFN in FM might be better defined considering that there is heterogeneity in this feature; for instance, it could be speculated that small-fiber damage represents an initiating event in some (but not all) patients with FM. Alternatively, it is possible that some patients in whom FM is diagnosed are instead affected by SFN with clinical features mimicking FM. Indeed, widespread pain and sensory symptoms can be a feature of non–length-dependent SFN (7,8). Muscle pain, a main symptom of patients with FM, may also be present in SFN, possibly related to involvement of intramuscular nerve fibers (9). Thus, SFN recently has been implicated in patients whose main symptoms are cramps and myalgia (10). Caro and Winter also evaluated whether degeneration of small nerve fibers occurred in a length-dependent pattern and concluded that although the loss of small nerve fibers was diffuse, distal sites were preferentially affected. Again, this should be reassessed in individual cases. Because the overall mean ⫾ SD thigh-to-calf ENFD ratio was 1.9 ⫾ 1.0, it is evident that, in at least some patients, proximal fiber loss was greater than, or at least equal to, distal loss, configuring a non–length-dependent pattern more compatible with the widespread pain symptoms of FM. This contrasts with the clinical findings of a “stocking distribution” and diminished pinwheel and vibratory perception in all patients. The relevance of these data is arguable, however, because the data were not precisely quantified, and considerable variability of responses (especially for vibration sense [11]) is usually observed in normal subjects. Although patients with conditions potentially causing polyneuropathy were excluded, 9 patients with FM also had rheumatoid arthritis, which is a condition that is significantly associated with neuropathy (12). In addition, the presence of a relevant number of cases of rheumatoid arthritis in this series could entail a bias with regard to the immunologic findings, e.g., the inverse correlation between serum IL-2R levels and calf ENFD. Franco Gemignani, MD University of Parma Parma, Italy 1. Caro XJ, Winter EF. Evidence of abnormal epidermal nerve fiber density in fibromyalgia: clinical and immunologic implications. Arthritis Rheumatol 2014;66:1945–54. 2. Uceyler N, Zeller D, Kahn AK, Kewenig S, Kittel-Schneider S, Schmid A, et al. Small fibre pathology in patients with fibromyalgia syndrome. Brain 2013;136:1857–67. 3. Oaklander AL, Herzog ZD, Downs HM, Klein MM. Objective evidence that small-fiber polyneuropathy underlies some illnesses currently labeled as fibromyalgia. Pain 2013;154:2310–6. 4. Giannoccaro MP, Donadio V, Incensi A, Avoni P, Liguori R. Small nerve fiber involvement in patients referred for fibromyalgia. Muscle Nerve 2014;49:757–9. 5. Serra J, Collado A, Sola R, Antonelli F, Torres X, Salgueiro M, et al. Hyperexcitable C nociceptors in fibromyalgia. Ann Neurol 2014;75:196–208. 6. Phillips K, Clauw DJ. Central pain mechanisms in chronic pain states—maybe it is all in their head. Best Pract Res Clin Rheumatol 2011;25:141–54.
7. Gorson KC, Herrmann DN, Thiagarajan R, Brannagan TH, Chin RL, Kinsella LJ, et al. Non-length dependent small fibre neuropathy/ganglionopathy. J Neurol Neurosurg Psychiatry 2008; 79:163–9. 8. Gemignani F, Giovanelli M, Vitetta F, Santilli D, Bellanova MF, Brindani F, et al. Non-length dependent small fiber neuropathy: a prospective case series. J Peripher Nerv Syst 2010;15:57–62. 9. Lauria G. Small fibre neuropathies. Curr Opin Neurol 2005;18: 591–7. 10. Lopate G, Streif E, Harms M, Weihl C, Pestronk A. Cramps and small-fiber neuropathy. Muscle Nerve 2013;48:252–5. 11. Martina IS, van Koningsveld R, Schmitz PI, van der Meche FG, van Doorn PA, for the European Inflammatory Neuropathy Cause and Treatment (INCAT) group. Measuring vibration threshold with a graduated tuning fork in normal aging and in patients with polyneuropathy. J Neurol Neurosurg Psychiatry 1998;65:743–7. 12. Agarwal V, Singh R, Wiclaf, Chauhan S, Tahlan A, Ahuja CK, et al. A clinical, electrophysiological, and pathological study of neuropathy in rheumatoid arthritis. Clin Rheumatol 2008;27: 841–4.
DOI 10.1002/art.38815
Reply To the Editor: We thank Dr. Gemignani for his scholarly comments and agree with his assessment that the pain in FM probably arises, in the main, from a peripheral neuropathic lesion. Additionally, our data (and everyday experience) suggest that this injury is likely to be immune mediated. He also observes, however, that we reported only descriptive statistics for our study populations and opines on the value of reporting the ENFD values of individual study participants. Such detailed reporting might have been ideal, but the reviewers dissuaded us from our original plan to report individual thigh–to-calf ENFD ratios, in a figure format, due to space limitations. Even in Dr. Gemignani’s previous study of non–length-dependent SFN (1), ENFD values were reported for only 8 of 44 affected patients, presumably for the same reason. In our report, however, we did show individual ENFD values for patients with FM and control subjects, graphed against age at the time of biopsy (see Figures 3 and 4). After data quality analysis, we chose to analyze and report the ENFD results as a continuous rather than a categorical variable, thus avoiding the arbitrary designations of “normal” or “abnormal.” This approach was suggested by the distribution of ENFD loss among our patients with FM, as evidenced by inspection of the spread along the y-axis in Figures 3 and 4. Following the traditional statistical approach, we also calculated values for skewness and kurtosis in our population, and the results demonstrated their relatively close Gaussian distribution (i.e., a normal, bell-shaped configuration). Had the data shown a more nonhomogeneous form, we might have been more inclined to class the outcomes categorically. Instead, our results suggest a smooth continuum of nerve fiber loss arising from a progressively severe small nerve injury due to a single causative agent. Despite these considerations, however, we too have occasionally observed an FM patient with a reversal of the
2. Stylianou E, Saklatvala J. Interleukin-1. Int J Biochem Cell Biol 1998;30:1075–9. 3. Tang C, Sula MJ, Bohnet S, Rehman A, Taishi P, Krueger JM. Interleukin-1 induces CREB-binding protein (CBP) mRNA in brain and the sequencing of rat CBP. Brain Res Mol Brain Res 2005;137:213–22. 4. Sambol EB, Ambrosini G, Geha RC, Kennealey PT, DeCarolis P, O’Connor R, et al. Flavopiridol targets c-KIT transcription and induces apoptosis in gastrointestinal stromal tumor cells. Cancer Res 2006;66:5858–66.
DOI 10.1002/art.38817
Reply To the Editor: We are appreciative that Chung took the time to carefully read our article and bring up interesting and salient points for discussion. We agree that some genes do not seem to “fit” with the conclusion that flavopiridol inhibits the induction of inflammatory mediator genes. However, we believe that we accurately represented the data by stating that most genes were not induced by IL-1 in the presence of flavopiridol, and that we were transparent by also showing genes that were exceptions and by including the raw data as supplemental information. In our study, the response of some genes was not expected, especially the responses of CCR5 and STAT1, both of which are implicated in inflammatory arthritis and osteoarthritis (1–3). For example, CDK-9 inhibition in the presence of interferon-␥ inhibited STAT1 activation and downstream effects in other cell types (4). It would be reasonable to expect a similar response to IL-1. Although messenger RNA levels (e.g., of STAT1 and CCR5) may not always correspond to protein levels or biologic activity, protein levels or biologic activities were not examined, because that was beyond the scope of the current study. We agree that further analysis of these genes would better explain the exact relationship of IL-1 and CDK-9/flavopiridol with these particular inflammatory mediators, especially in the context of cartilage and chondrocytes. There are several additional points to consider when evaluating these unexpected findings. First, in our experiments we used a 5-hour time point for the PCR array analysis. This time point was chosen because it yielded high-magnitude responses to the IL-1 treatment, but it does not necessarily preclude the possibility of secondary responses that are not regulated in the same CDK-9–dependent manner. Second, primary human cell strains and human cartilage explants remain largely untested with respect to CDK-9/flavopiridol and the regulation of primary response gene transcriptional elongation. These cells and tissues may not respond exactly as the other cell lines or tissues in the studies referenced by Chung. Finally, there is considerable variation in the response of human cartilage and isolated chondrocytes to IL-1. In our study, we repeated all experiments on samples from least 3 different human donors, indicating that the results shown are somewhat reproducible for both the expected responses and the unexpected responses of genes such as CCR5 and STAT1.
LETTERS
In summary, most of the inflammatory genes in the PCR array panel responded similarly, although there were a few outliers that had unexpected responses. Given the important roles of some of these outliers in arthritis, their relationship with the CDK-9–flavopiridol axis is fertile ground for further investigations. Dominik R. Haudenschild, PhD Jasper H. N. Yik, PhD University of California, Davis Sacramento, CA 1. Kasperkovitz PV, Verbeet NL, Smeets TJ, van Rietschoten JG, Kraan MC, Kraan T, et al. Activation of the STAT1 pathway in rheumatoid arthritis. Ann Rheum Dis 2004;63:233–9. 2. Lee YH, Bae SC, Song GG. Association between the chemokine receptor 5 ⌬32 polymorphism and rheumatoid arthritis: a metaanalysis. Mod Rheumatol 2013;23:304–10. 3. Qin SX, Rottman JB, Myers P, Kassam N, Weinblatt M, Loetscher M, et al. The chemokine receptors CXCR3 and CCR5 mark subsets of T cells associated with certain inflammatory reactions. J Clin Invest 1998;101:746–54. 4. Terashima T, Hague A, Kajita Y, Takeuchi A, Nakagawa T, Yokochi T. Flavopiridol inhibits interferon-␥-induced nitric oxide production in mouse vascular endothelial cells. Immunol Lett 2012;148:91–6.
DOI 10.1002/art.38812
Fibromyalgia syndrome and small-fiber neuropathy: comment on the article by Caro and Winter To the Editor: I read with interest the recent article by Caro and Winter (1), in which they reported decreased epidermal nerve fiber density (ENFD) in a series of 41 patients with fibromyalgia (FM), in accordance with other recent studies suggesting the involvement of small nerve fibers in FM as shown by reduced ENFD (2–4) and abnormal function of silent C-fiber nociceptors (5). These findings strengthen the concept that pain in FM syndrome is actually of the neuropathic type and point to the role of peripheral components, apparently opposite to the current opinion of FM as a condition characterized by central pain amplification (6). Unfortunately, Caro and Winter reported only that the overall mean ENFD was lower in patients compared with controls, without detailing the findings in individual cases. This is a crucial point, because the involvement of small nerve fibers seems to be limited to a subset of patients with FM. In previous studies by Oaklander et al (3) and Giannoccaro et al (4), only a minority of patients had reduced intraepidermal ENFD in ¨ ¸ceyler et al (2) reported abnormalities skin biopsy specimens. U in the function and morphology of small fibers in all 25 FM patients they studied. However, direct evidence of damage to small fibers was provided by reduced intraepidermal ENFD in only 16 patients, whereas the remaining patients had abnormal results on other tests (quantitative sensory testing and pain-related evoked potentials) that are not strictly specific for small fibers but rather investigate the whole length of the somatosensory system. In contrast, Serra et al (5) demon-
LETTERS
3527
strated hyperexcitability of silent (type 1B) C-fiber nociceptors, resembling that in small-fiber neuropathy (SFN), in the majority of patients with FM (23 of 30). The role of SFN in FM might be better defined considering that there is heterogeneity in this feature; for instance, it could be speculated that small-fiber damage represents an initiating event in some (but not all) patients with FM. Alternatively, it is possible that some patients in whom FM is diagnosed are instead affected by SFN with clinical features mimicking FM. Indeed, widespread pain and sensory symptoms can be a feature of non–length-dependent SFN (7,8). Muscle pain, a main symptom of patients with FM, may also be present in SFN, possibly related to involvement of intramuscular nerve fibers (9). Thus, SFN recently has been implicated in patients whose main symptoms are cramps and myalgia (10). Caro and Winter also evaluated whether degeneration of small nerve fibers occurred in a length-dependent pattern and concluded that although the loss of small nerve fibers was diffuse, distal sites were preferentially affected. Again, this should be reassessed in individual cases. Because the overall mean ⫾ SD thigh-to-calf ENFD ratio was 1.9 ⫾ 1.0, it is evident that, in at least some patients, proximal fiber loss was greater than, or at least equal to, distal loss, configuring a non–length-dependent pattern more compatible with the widespread pain symptoms of FM. This contrasts with the clinical findings of a “stocking distribution” and diminished pinwheel and vibratory perception in all patients. The relevance of these data is arguable, however, because the data were not precisely quantified, and considerable variability of responses (especially for vibration sense [11]) is usually observed in normal subjects. Although patients with conditions potentially causing polyneuropathy were excluded, 9 patients with FM also had rheumatoid arthritis, which is a condition that is significantly associated with neuropathy (12). In addition, the presence of a relevant number of cases of rheumatoid arthritis in this series could entail a bias with regard to the immunologic findings, e.g., the inverse correlation between serum IL-2R levels and calf ENFD. Franco Gemignani, MD University of Parma Parma, Italy 1. Caro XJ, Winter EF. Evidence of abnormal epidermal nerve fiber density in fibromyalgia: clinical and immunologic implications. Arthritis Rheumatol 2014;66:1945–54. 2. Uceyler N, Zeller D, Kahn AK, Kewenig S, Kittel-Schneider S, Schmid A, et al. Small fibre pathology in patients with fibromyalgia syndrome. Brain 2013;136:1857–67. 3. Oaklander AL, Herzog ZD, Downs HM, Klein MM. Objective evidence that small-fiber polyneuropathy underlies some illnesses currently labeled as fibromyalgia. Pain 2013;154:2310–6. 4. Giannoccaro MP, Donadio V, Incensi A, Avoni P, Liguori R. Small nerve fiber involvement in patients referred for fibromyalgia. Muscle Nerve 2014;49:757–9. 5. Serra J, Collado A, Sola R, Antonelli F, Torres X, Salgueiro M, et al. Hyperexcitable C nociceptors in fibromyalgia. Ann Neurol 2014;75:196–208. 6. Phillips K, Clauw DJ. Central pain mechanisms in chronic pain states—maybe it is all in their head. Best Pract Res Clin Rheumatol 2011;25:141–54.
7. Gorson KC, Herrmann DN, Thiagarajan R, Brannagan TH, Chin RL, Kinsella LJ, et al. Non-length dependent small fibre neuropathy/ganglionopathy. J Neurol Neurosurg Psychiatry 2008; 79:163–9. 8. Gemignani F, Giovanelli M, Vitetta F, Santilli D, Bellanova MF, Brindani F, et al. Non-length dependent small fiber neuropathy: a prospective case series. J Peripher Nerv Syst 2010;15:57–62. 9. Lauria G. Small fibre neuropathies. Curr Opin Neurol 2005;18: 591–7. 10. Lopate G, Streif E, Harms M, Weihl C, Pestronk A. Cramps and small-fiber neuropathy. Muscle Nerve 2013;48:252–5. 11. Martina IS, van Koningsveld R, Schmitz PI, van der Meche FG, van Doorn PA, for the European Inflammatory Neuropathy Cause and Treatment (INCAT) group. Measuring vibration threshold with a graduated tuning fork in normal aging and in patients with polyneuropathy. J Neurol Neurosurg Psychiatry 1998;65:743–7. 12. Agarwal V, Singh R, Wiclaf, Chauhan S, Tahlan A, Ahuja CK, et al. A clinical, electrophysiological, and pathological study of neuropathy in rheumatoid arthritis. Clin Rheumatol 2008;27: 841–4.
DOI 10.1002/art.38815
Reply To the Editor: We thank Dr. Gemignani for his scholarly comments and agree with his assessment that the pain in FM probably arises, in the main, from a peripheral neuropathic lesion. Additionally, our data (and everyday experience) suggest that this injury is likely to be immune mediated. He also observes, however, that we reported only descriptive statistics for our study populations and opines on the value of reporting the ENFD values of individual study participants. Such detailed reporting might have been ideal, but the reviewers dissuaded us from our original plan to report individual thigh–to-calf ENFD ratios, in a figure format, due to space limitations. Even in Dr. Gemignani’s previous study of non–length-dependent SFN (1), ENFD values were reported for only 8 of 44 affected patients, presumably for the same reason. In our report, however, we did show individual ENFD values for patients with FM and control subjects, graphed against age at the time of biopsy (see Figures 3 and 4). After data quality analysis, we chose to analyze and report the ENFD results as a continuous rather than a categorical variable, thus avoiding the arbitrary designations of “normal” or “abnormal.” This approach was suggested by the distribution of ENFD loss among our patients with FM, as evidenced by inspection of the spread along the y-axis in Figures 3 and 4. Following the traditional statistical approach, we also calculated values for skewness and kurtosis in our population, and the results demonstrated their relatively close Gaussian distribution (i.e., a normal, bell-shaped configuration). Had the data shown a more nonhomogeneous form, we might have been more inclined to class the outcomes categorically. Instead, our results suggest a smooth continuum of nerve fiber loss arising from a progressively severe small nerve injury due to a single causative agent. Despite these considerations, however, we too have occasionally observed an FM patient with a reversal of the
